Return to Sender: How to Successfully Address the Cable Industry's Use of Inaccurate Zip+4 Data in Effective Competition Proceedings

By
Stephen J. Guzzetta, Richard D. Treich & Heidi Arnson

Over the past several years, cable operators have filed numerous Petitions for Special Relief with the Federal Communications Commission (the “FCC”) seeking determinations of effective competition in communities across the country and the concomitant elimination of local rate regulation.[1] In most cases, operators attempt to prove the existence of effective competition in a particular franchise area through the use of the “competing provider” test set forth in Section 623(l)(1)(B)(i)-(ii) of the Cable Communications Policy Act of 1984, as amended, 47 U.S.C. § 543(l)(1)(B)(i)-(ii) (the “Cable Act”).[2] This entails showing that: (i) the franchise area(s) at issue are served by at least two unaffiliated multichannel video programming distributors each of which offers comparable programming to at least 50 percent of the households in the franchise area; and (ii) the number of households subscribing to multichannel video programming services offered by multichannel video programming distributors other the largest multichannel video programming distributor exceeds 15 percent of the occupied households in the relevant franchise area(s).[3]

Based on the almost ubiquitous availability of direct broadcast satellite (“DBS”) service, and the FCC’s consistent interpretation of Section 623(i)(1)(B)(i), the first prong of the competing provider test is typically satisfied in most cases and generally is not worth investigating and challenging. The second prong, however, is where factual disputes with cable operators usually arise and where local rate regulation authorities should devote their time and resources because it is possible to evaluate and successfully challenge the accuracy and reliability of a cable operator’s DBS penetration calculations. Those calculations typically involve dividing the number of DBS subscribers located in designated franchise area(s) (as obtained from the Satellite Broadcasting and Communications Association (the “SBCA”)) by the number of occupied households in the franchise area(s) at issue (which cable operators usually obtain from 2000 U.S. Census reports). Thus, the two variables that need to be tested in an archetypal effective competition proceeding are the cable operator’s reported DBS subscriber and occupied household figures. This article will focus on issues associated with accurately identifying the number of DBS subscribers in a franchise area.

Traditionally, the cable industry obtained DBS subscriber numbers for effective competition filings by providing five-digit zip code information to the SBCA, which would then quantify the number of DBS subscribers associated with specified five-digit zip codes. Over time, however, it has become apparent that 5-digit zip codes used by the cable industry do not always carefully follow the territorial boundaries of a franchise area. As a consequence, the DBS subscriber numbers obtained from the SBCA and submitted to the FCC by cable operators would frequently overstate or understate the true number of DBS subscribers in the franchise area(s) at issue (thereby skewing the true DBS penetration rate in a given franchise area). To address this problem, the cable industry began using a complex and untested DBS subscriber allocation methodology developed by Media Business Corp. (“MBC”) to estimate the number of DBS subscribers in a particular franchise area. The complexity built in to MBC’s approach introduces numerous potential sources of error, both in terms of assumptions, data inputs and data manipulation. Despite the infirmities inherent in MBC’s calculations, and notwithstanding the fact that MBC’s results have never been thoroughly evaluated for accuracy (largely because cable operators will not make underlying data available for review), the FCC routinely and perfunctorily approved MBC’s subscriber allocation methodology over the objections of local rate regulation authorities.[4] As of late, however, the FCC appears to be viewing cable operators’ DBS subscriber penetration calculations with more skepticism. For example, in light of some absurd cases in which video subscription rates exceeded 100 percent in a franchise area,, the FCC decided in 2008 to require cable operators to use more accurate zip+4 zip codes (e.g., 80104-7758) in obtaining DBS subscriber reports from the SBCA, at least for zip codes that overlap different franchise areas.[5] While the FCC has now retreated from this position, it has clearly pronounced that it will “dismiss evidence that shows obviously inaccurate or highly questionable levels of subscription regardless of the format of such evidence. The Commission suggests that a cable operator with Five-Digit Zip Code evidence showing such flawed results consider the use of Nine-Digit Zip Code data or some other kind of evidence that credibly sustains its burden of proof . . .”[6]

A case study of the Petition for Special Relief (the “Petition”) Comcast of Minnesota, Inc. (“Comcast”) filed with the North Metro Telecommunications Commission[7] (the “NMTC”) illustrates the deficiencies in the MBC subscriber allocation methodology and shows that even the use of zip+4 zip codes can produce inaccurate DBS subscriber figures if a cable operator does not confirm that all the zip+4 zip codes it supplies to the SBCA are actually located within the territorial boundaries of a specific franchise area. As in most cases, Comcast’s Petition initially relied on the MBC methodology to calculate the DBS subscriber totals for the NMTC’s individual Member Cities. After reviewing the Petition, the NMTC discovered the DBS subscriber numbers produced using the MBC methodology were extremely inaccurate. By way of example, the NMTC found that Comcast and MBC overestimated DBS subscribership in Blaine by 16.31% and in Centerville by 34.99%. Additionally, the NMTC determined that for three of the NMTC’s Member Cities, the MBC methodology mistakenly included seven (7) zip codes (and associated DBS subscribers) that fall outside the three Member Cities and inadvertently missed eight (8) zip codes (and associated DBS subscribers) that lie within the boundaries of the three Member Cities. Clearly, then, MBC’s DBS subscriber allocation methodology cannot be considered accurate or trustworthy.

In its Reply to the NMTC’s Opposition, Comcast filed completely new DBS subscriber data for the Member Cities using zip+4 zip codes. Comcast, however, did not explain how it identified the zip+4 zip codes applicable to the Member Cities or how it verified that the individual zip+4 zip codes it supplied to the SBCA did not overlap individual Member Cities or fall entirely outside the boundaries of the Member Cities. In any event, the zip+4 DBS subscriber data included in Comcast’s Reply were very different from the DBS subscriber numbers derived from the MBC methodology. Table 1 summarizes those differences and computes the effective error rate.

Table 1

LN

Community
Allocated DBS Subs by
MBC
DBS Subs by
Zip Code Plus 4

Error Rate
1
Blaine
5,050
4,420
-12.5%
2
Centerville
490
334
-31.8%
3
Circle Pines
454
312
-31.3%
4
Ham Lake
1,364
1,358
-0.4%
5
Lexington
226
129
-42.9%
6
Lino Lakes
1,461
1,525
4.4%
7
Spring Lake Park
521
506
-2.9%
8
Total
9,566
8,584
-10.3%

At least with respect to the Member Cities, Comcast’s updated subscriber data clearly impugns the reliability of MBC’s DBS subscriber allocation calculations.

In order to assess the accuracy of Comcast’s revised DBS subscriber penetration rates for the Member Cities, the NMTC investigated the DBS subscriber figures Comcast proffered for Lexington and Circle Pines using zip+4 zip code data, as the penetration rates in those communities were just above the 15% level required by the competing provider test. The NMTC’s investigation began by gathering the property tax records for all properties located in Lexington and Circle Pines using official Anoka County, Minnesota records (which contain property addresses and are used to allocate property tax payments to municipalities). Unfortunately, the property tax records maintained by Anoka County only contain a 5-digit zip code for each property address. However, the NMTC was able to find a consulting company that could assign a zip+4 zip code to each address and 5-digit zip code contained in the Anoka County property records for Lexington and Circle Pines. The NMTC was then able to compare its list of zip+4 zip codes against the lists Comcast sent to the SBCA.

The NMTC’s analysis of the two lists revealed that Comcast had erroneously included 97 zip+4 zip codes in its Reply that were not associated with any households in Lexington (and incorrectly excluded 17 zip+4 zip codes). As a result, Comcast mistakenly reported that there were 129 DBS subscribers in Lexington instead of 88 subscribers (an error rate of 47%). When the correct number of DBS subscribers Lexington was plugged in to the FCC’s approved DBS subscriber penetration rate formula (along with correct occupied household data),[8] the actual DBS subscriber penetration rate fell well below 15%, as indicated in Table 2.

Table 2
Lexington, MN
LN
Item
DBS Subscribers
Occupied Households
DBS Penetration
1
Comcast Original Petition
226
847
26.68%
2
Comcast Reply
129
804
16.04%
3
NMTC Surreply
88
804
10.95%
4
Error - Original
-156.82%


5
Error - Reply
-46.59%



In light of the compelling evidence the NMTC submitted to the FCC, Comcast withdrew its request for a finding of effective competition in Lexington. This shows local rate regulation authorities can successfully oppose an effective competition petition if they are (i) willing to devote adequate time and resources to analyzing a cable operator’s DBS penetration calculations and (ii) have access to reliable sources of data that can be used to verify the cable operator’s DBS subscriber and/or occupied household data.

[1] See 47 C.F.R. §§ 76.7 and 76.907 (2008). In addition to eliminating local authority over basic service, installation and associated equipment rates, a finding of effective competition will free cable operators from federal uniform rate requirements and tier buy-through prohibitions. See, e.g., 47 C.F.R. §§ 921 and 984 (2008).
[2] See also 47 C.F.R. § 76.905(b)(2)(i)-(ii) (2008).
[3] 47 U.S.C. § 543(l)(1)(B)(i)-(ii) and 47 C.F.R. § 76.905(b)(2)(i)-(ii) (2006).
[4] See, e.g., In the Matter of Adelphia Communications: Petition for Determination of Effective Competition in Various North Carolina Communities, Memorandum Opinion and Order, 22 FCC Rcd 4458 (2007).
[5] See, e.g., Public Notice, Commission Revises and Suspends Pending OMB Approval New Standards For Showings of Effective Competition for Cable Service, 23 FCC Rcd 13252 (2008).
[6] Public Notice, Commission Clarifies Standards for Evidence of Competing Provider Effective Competition for Cable Service, DA 09-1361, 2009 FCC LEXIS 2993 (Rel. June 18, 2009).
[7] The NMTC member municipalities are Blaine, Centerville, Circle Pines, Ham Lake, Lexington, Lino Lakes, and Spring Lake Park, Minnesota (collectively, the “Member Cities”).
[8] The NMTC did challenge Comcast’s 8-year old occupied household data by utilizing official 2007 occupied household figures produced by the Metropolitan Council in the Twin Cities area.